LF: Re: GPS Coherent PSK Transmission

["Talbot Andrew" <ACTALBOT@dstl.gov.uk>]

Yes, but if you'd seen previous EMails you would have realised I must use this frequency for other reasons.

 

G3PLX, who prompted these PSK transmissions, is developing a very simple cheap and cheerful way of locking a receiver to the GPS 1PPS pulse, and at the moment he can only listen on frequencies that are equal to ( N + 1 / M) Hz   where N and M are integers.  Using my DDS source driven from the 5MHz frequency standard, the only frequency in the entire band that meets this criteria is 5MHz * 7 / 256  = 136718.75.    Arbitrary frequencies could have a frequency error as high as 561uHz, which is far too high for these tests at 4E-9 error - we need long term (several hours) accuracies of just a few parts in 10^-10

 

So as far as we're concerned bandplans can go to the wall when this sort of requirement rears its head !

 

I hardly think a sub 0dBm ERP transmission is going to cause a problem to those wideband incoherent signallers !

 

A parallel longer term solution I'm looking at for this signalling is a standard locked to GPS with a frequency that is more 'DDS friendly'    A PLL with a 1Hz comparison frequency will enable a 4.194304MHz oscillator to be locked which with a 32 bit DDS will give all frequencies at multiples of  1/1024 Hz exactly.   A big difference between traditional frequency standards and the requirements for low speed signalling here, is that the short term phase noise is less important.   Where each PSK symbol is 30 seconds long, then phase jitter every second gets integrated out.   If such a standard were used for HF to UHF then this would result in an unacceptable frequency blip.  Another option for better phase noise will be to lock a 4.096MHz oscillator to the 5MHz standard (using 8kHz reference).  Tis is not so good on a DDS but will give exact 1 Hz values every 125Hz, and so quarter Hz values every 31.25Hz  - so I'm sure a few frequencies can be found in the proper part of the band with these multiples.

 

Andy  'JNT

-----Original Message-----
From: G0MRF@aol.com [mailto:G0MRF@aol.com]

In a message dated 1/6/02 5:11:10 PM GMT Standard Time, G4JNT@thersgb.net writes:
sequence is a '0'.  Frequency 136718.75000 Hz

Isn't the data transmission 'slot' above 137.3  ish  ??

-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.